Identifying stakeholder perspectives and worldviews on sustainable urban tourism development using a Q-sort methodology

Tourism growth, particularly in cities, is coming under increased scrutiny. However, even often visited cities appear to find it difficult to agree upon a strategy to limit tourism growth. The current paper investigates this issue by looking at the extent to which different stakeholders’ perspectives on tourism development align. Q-sort methodology is employed to find the main worldviews and the extent to which they are shared by stakeholders in similar roles (e.g. policymakers, industry, resident). Results point to the existence of five different worldviews, which differ in the extent to which tourism growth is desirable or problematic and whether resident participation is advantageous or counterproductive. Stakeholders have highly different worldviews, even those with similar roles, which may help explain the difficulty to change the tourism growth paradigm as they limit opportunities for generating new consensus-based collective solutions. If we accept that tourism development strategies are driven and informed at least in part by individual worldviews, it may be impossible to make ‘objective’ policy choices. Instead, it might be more useful to explore possibilities to allow stakeholders to express their worldviews to better understand what sustainable tourism development entails for different people at different places and moments in time. ARTICLE HISTORY Received 30 April 2019 Accepted 21 January 2020

Identifying stakeholder perspectives and worldviews on sustainable urban tourism development using a Q-sort methodology

Introduction
In the overtourism debate that has come to dominate discussions on urban tourism in recent years, most research up to now has been dedicated to better appreciating the perceived impacts and underlying causes of these impacts (see e.g. Fava & Rubio, 2017;Koens, Postma, & Papp, 2018;Novy, 2018) and management strategies to deal with overtourism (see e.g. Pechlaner, Innerhofer, & Erschbamer, 2019;UNWTO, 2018). One of the main outcomes of these efforts seems to be a wide recognition of the gap between (economic) tourism development on the one hand and the perceived needs and desires of local city users on the other. Another issue that is commonly brought up, is the apparent lack of leadership and coordination when it comes to managing and regulating tourism development (Koens, Postma, & Papp, 2019), echoing similar findings in earlier research (Dodds, 2007;Timur & Getz, 2009).
In fact, some argue that it is striking how the (official) discourse with regards to tourism in most cities remains limited to a focus on growth, while aiming to mitigate possible negative effects of tourism (Aall, Dodds, Saelensminde, & Brendehaug, 2015;Koens, Postma and Papp, 2019). Even in Copenhagen, which is widely regarded as being a front runner on the topic of socially inclusive tourism development, growth is accepted as a given (Wonderful Copenhagen, 2017). In contrast, the (media) discourse in the context of overtourism can be considered anti-tourism as it strongly denounces tourism development, with little attention to the different causes and levels of perceived overtourism (Clancy, 2019). This rather binary way of framing ignores the different perspectives on appropriate reference points for urban tourism governance that numerous stakeholders 'on the ground' hold and (try to) express in discussions on urban tourism planning. This limits the possibilities and/or implementation of alternative management strategies, let alone the radical systemic changes that are actually required from a long-term sustainable urban development perspective (Koens, Melissen, Mayer, & Aall, 2019). To better understand the mechanisms at work here and their implications for ultimate decision-making on the ground, it is therefore necessary to better understand these various perspectives towards tourism development.
This paper explores particular perspectives, or worldviews, of relevant stakeholders that (appear to) influence current (and future) urban tourism governance, in five European cities. The relevance of uncovering and understanding such worldviews is that they can serve as alternative frames to mediate the discussion in a way that is more productive than the two discourses that currently dominate the debate, as will be further illustrated in the next section of this paper. Subsequently, the specific approach used in this project to establish current worldviews held by relevant stakeholders in these cities in relation to urban tourism development is described in detail. The results obtained through applying this so-called mixed method Q methodology are then presented and discussed in the remainder of this paper. This includes an appreciation of the influence of the role or function of stakeholders in a city on their worldview as well as the possible implications for exploring and implementing sustainable urban tourism development strategies.

The potential impact of worldviews on destination management
Until the 1990s, urban tourism policymaking and destination development was mainly the responsibility of and performed by local government or governmental institutions (e.g. government-led Destination Management Organizations (DMOs)). Even though there were concerns regarding the actual impacts of tourism (Van Der Borg, 1992), there was relative clarity with regards to problem definition as well as the identification of solutions and/or actions (Van de Riet, 2003). Over the course of the past thirty years however, in many cities government has relinquished much of its control as it has downsized and spread responsibilities and resources across a wide range of stakeholders, largely driven by political and societal changes (Dredge, 2006). With the power of government seemingly reduced, current urban tourism governance systems rely increasingly on local tourism policy networks. In these networks, different actors need to work together to achieve coherent destination management (decision-making and policies), which has made destination management increasingly complex. Differences in expertise, experience, knowledge and worldviews can result in policy problems and potential solutions being perceived quite differently, not least because involved actors oftentimes have conflicting interests (Nieuwland & van Melik, 2018;Pechlaner, Kozak, & Volgger, 2014;Phi, Dredge, & Whitford, 2014).
Tourism planning literature more and more suggests a specific strategy to cope with these issues: to focus on collaborative processes, which include a broad involvement of diverse groups of stakeholders in urban tourism governance. Collaborative planning is argued to contribute to effective policymaking, particularly if it leads to consensus and shared perspectives among stakeholders (see. e.g. d 'Angella & Go, 2009;Healey, 2003). Simultaneously, collaborative planning has been criticized for perpetuating the hegemony of local elites, as well as inequalities between stakeholders, through a focus on consensus seeking (Beritelli, 2011b).
In addition, an emphasis on explicit plans and strategies, as well as the fact that in practice perspectives are still commonly discussed at the level of stakeholder groups and decision-making is often based on the implicit assumption that representatives of such stakeholder groups voice the opinion of all members of that group, rather than at an individual level (see e.g. Austrian Hotelier Association & Roland Berger, 2018;Byrd, 2007;Jamal & Stronza, 2009;Ritchie & Crouch, 2003;Ruhanen, 2009;UNWTO, 2018;Wang & Fesenmaier, 2007;Zehrer & Hallmann, 2015). This limits the opportunities for less well-organized and less powerful organizations and individuals to engage in ultimate decision-making, even when some of these actors are more likely to come up with the radical and innovative solutions that are actually required (Baggio & Cooper, 2010). In other words, many current practices that are aimed at collaboration and stakeholder involvement actually limit the possibilities for conflicting perspectives or worldviews of (groups of) individuals to be taken into account (Beritelli, 2011b). This is particularly problematic given the recent developments in urban tourism, where the overtourism debate has resulted in highly polarised conflicting perspectives among stakeholder groups (Koens et al., 2018). While solutions for overcoming the issues with current governance practices have been proposed (see e.g. Frantzeskaki, Broto, Coenen, & Loorbach, 2017;Pechlaner et al., 2014), so far it has proven difficult to overcome these problems, especially on the ground.
In this paper, we use the concepts of individual and collective worldviews to better understand the complexity that urban tourism destinations face in recognizing, accounting for and benefitting from varying stakeholder perspectives on destination development and management, in the context of sustainable urban tourism development and overtourism. As such, the paper contributes to the limited body of work on this topic (Lew, 2017;Seyfi, Michael Hall, & Fagnoni, 2018) and could contribute to improving urban tourism governance practices on the ground.
Worldviews can be defined as the ways in which individuals and groups of individuals define their taken-for-granted lifeworld and maintain it through world-making actions (Hollinshead, Ateljevic, & Ali, 2009;Lew, 2017). Different worldviews will likely result in different preferences and actions with regards to sustainable urban tourism development. A distinction can be made between individual and collective worldviews, with the former being a personal perspective and the latter the explicit worldview of a (group of) organization(s). The importance of collective worldviews is easier to recognize in that they shape the dominant discourse with regards to a topic or debate (Johnstone, 2018). Within tourism, for example, the rise of overtourism has been attributed to the dominance of neoliberal collective worldviews with regards to and of those 'in charge of' tourism development (Russo & Scarnato, 2018). While specific local discourses may differ, depending on the history and characteristics of a specific destination, the dominant discourse in many destinations seems to (have) be(en) one that favours economic growth through tourism with relatively limited regulation. The emphasis in many destinations on enabling free markets to thrive, with the aim of tourism growth, seems to have become even stronger after the economic crisis of 2008 as growth in other industries faltered (Joppe, 2018;Koens, Postma and Papp, 2019). Policy recommendations that emphasize regulation and control to limit tourism growth did not fit within this discourse and, as such, were rarely acted upon.
Although long-standing dominant collective worldviews can become incorporated into personal worldviews over time, collective worldviews are commonly not shared by everyone and individuals can display a different worldview from the one advocated by the organizations they work for or stakeholder group they are deemed to belong to in traditional stakeholder engagement processes. Different worldviews may compete or develop as counternarratives or discourses under a dominant worldview (Hedlund-de Witt, de Boer, & Boersema, 2014). Taking the overtourism example, over time residents in major tourism destinations increasingly have developed feelings of disenfranchisement and disillusion with regards to tourism development. In Amsterdam and Barcelona this resulted in tourism serving as a central argument in municipal elections, with parties critical of tourism getting voted in power on the promise of stricter tourism regulation and control (Jakobs, 2018;Russo & Scarnato, 2018). While this might suggest the development of a new collective worldview, taming the 'continuous tourism growing machine' has actually proven difficult. In spite of political pressure and a more regulatory narrative among most tourism stakeholders, tourism continues to grow, and tourist numbers continue to increase, with tourism now expanding into new areas (Mansilla & Milano, 2019, p. 16).
Even though it is increasingly accepted that tourist destination communities consist of networks of individuals who choose to cooperate with others on the basis of interpersonal relationships, rather than rational logic or the organization they work for (Beritelli, 2011a;Pforr, 2006;Scott, Baggio, & Cooper, 2008), the role of personal worldviews has been relatively overlooked in academic research (Bramwell & Lane, 2014) and stakeholder engagement processes on the ground. However, such worldviews may shed light on some of the difficulties that stakeholders come across when (contributing to) developing long-term strategies to manage tourism, particularly with regards to efforts to apply collaborative planning principles and the associated 'debates and mediations among diverse actors on planning and future development' (Jóhannesson & Huijbens, 2010, p. 432). Interactions between people with similar personal worldviews can lead to implicitly shared meanings and understandings of desirable outcomes of policymaking, which can play a vital role when it comes to agenda setting or implementation of tourism projects and initiatives (Beritelli, 2011a). Dissimilar personal worldviews can seriously hamper agenda setting and implementation, which could very well play a role in explaining and dealing with the challenges several cities are currently experiencing in relation to developing and effectuating sustainable urban tourism development strategies.
A deeper understanding of relevant stakeholders' personal worldviews, not just among policymakers but also other stakeholders who participate in (current) collaborative governance systems, could very well clarify the experienced difficulties in coming to a shared understanding of what entails (planning for) sustainable urban tourism development.

Methodology
This paper reports on a research project dedicated to establishing current personal worldviews of relevant stakeholders of sustainable urban tourism development in five European cities through applying a Q methodology. Q methodology is an innovative mixed methods methodology that allows for the systematic exploration of the construction and sharing of representative viewpoints among different individuals (Hutson & Montgomery, 2010;Stergiou & Airey, 2011) and has been described as a 'scientific study of human subjectivity' (Goldman, 1999, p. 589). Even though it is used regularly to appreciate beliefs and values in relation to governance issues (Addams & Proops, 2000;Durning, 1999), Q methodology is little used in tourism studies with some notable exceptions (e.g. Huang, Qu, & Montgomery, 2017;Phi et al., 2014;Wijngaarden, 2017). The value of Q methodology is that it combines the mathematical rigour of quantitative approaches with the interpretative component that is common in qualitative research (Robbins & Krueger, 2000). The basic premise of the methodology is that participants are asked to rank and comment on a set of items or statements, from which shared or 'typical' perspectives are extracted by means of a rotated factor analysis. This allows for richer insights than can be gained from a pure quantitative approach, whilst also mitigating researcher bias in the interpretation of participants' narratives (Huang et al., 2017).
The first step within Q methodology involves the development of a 'concourse' that will form the basis of the actual Q sort. The concourse represents the range and scope of perspectives and communications about the topic under investigation. It results into a collection of approximately 30-60 relevant items or statements, which have been developed by means of extensive qualitative research (Newman & Ramlo, 2010). In the current research, interviews were performed in six European cities (Amsterdam, Belgrade, Darmstadt, Goteborg, Stavanger, Valencia) to identify and explore different views and perspectives on sustainable urban tourism development. Participants were purposefully chosen to reflect a diversity of perspectives. They included politicians, policymakers, employees of Destination Marketing Organizations (DMOs), residents and resident interest groups, owners and employees of tourism businesses and attractions (e.g. hotels, tour operators, museums), entrepreneurs, (technology) consultants and local academics. A total of 60 in-depth interviews were performed, by a team of interviewers. To ensure consistency, an interview guide was used, and application of this guide was practiced in a training that all interviewers attended. Interviews lasted between 45 and 60 minutes and were recorded and transcribed ad verbatim. The transcripts were analyzed using the qualitative data software NVivo 12. First, coding of key phrases and segments of text was performed, after which similar views and perspectives were grouped by means of a process of thematic analysis. The final step was to make statements from the interview data, mostly using direct quotes from the interviews. The interview statements were compared with the existing literature (e.g. Canavan, 2013;Getz, Timur, & Theobald, 2012) to also encapsulate previously found views and perspectives. This work resulted in an initial set of 300 statements that represented different viewpoints and perspectives on sustainable urban tourism development.
In research based on the Q methodology, the number of perspectives the researcher would like to uncover relates to the number of participants and statements. To determine the number of participants and statements, guidelines of Webler, Danielson, and Tuler (2009) were followed. As a general rule, for each perspective that one wants or expects to uncover one needs at least three participants to positively 'load' onto that particular perspective. The expected number of perspectives for this research was set at six, based on the initial assumption that the particular role or function of a stakeholder in a city would influence his or her worldview and assigning six key (main) roles/functions to prospective participantssee Table 1. Therefore, the minimum number of participants required was six perspectives x three people = 18 participants. However, 23 people participated, all purposively chosen based on their stakeholder profile (linked to the six suggested perspectives) and all originating from the same European cities as where the initial interviews were conducted. This automatically entails that the maximum number of statements to be used in this study was determined at 23 × 3 = 69 (Webler et al., 2009). Therefore, from the 300 initial statements, a pre-selection of 100 was made based on relevance to key impacts of urban tourism development, as established based on relevant literature and the analysis of the initial interviews, i.e. quality of life for residents, attractiveness of the city for tourists, equitability, perceived stakeholder involvement, ecological impact and economic impact, and by filtering out doublets, also by combining statements. These 100 statements were then further assessed and evaluated to see if they could be somehow combined without losing their representativeness of the concourse. This resulted in 70 statements which were then divided over six categories representing the six key impacts to check for each category having a similar number of statements and avoiding that a particular category was over-or underrepresented. The next step was to place these remaining statements into a 3 × 3 grid (following Dryzek & Berejikian, 1993;Jeffares & Skelcher, 2011). The grid regards discursive qualities relation, agency, and motivation on one axe and considers definitions, opinions or prescriptions on another. By assessing and rephrasing all statements by means of this grid, one obtains a well-represented and diverse set of statements, with different types of claims (Jeffares & Skelcher, 2011;Stergiou & Airey, 2011). The 70 statements were combined and rephrased to suit the description of the cells and this resulted in the final set of 60 statements. This amount fits well between the minimum of 46 (two participants loading on a perspective × 23 participants) and the maximum of 69 (3 × 23) statements (Webler et al., 2009). The number of statements determined the grid that was used in the actual Q sortssee Figure 1. The grid ranges from −6 (completely disagree), to 0 (neutral), to 6 (completely agree) and participants are asked to position each statement to a spot in the grid. All Q sorts were facilitated during face-to-face sessions by the same researcher, who visited the individual participants in their home city. The researcher also asked participants to elaborate on their reasoning for their individual Q sort and by explicitly asking them to reflect on the reasons for assigning particular statements to the −6, −5, +5 and +6 array positions in the grid. These responses were audio recorded.

Results
After conducting the Q sorts with the participants in the cities, the individual Q sorts as laid out by each of the participants were entered in the software PQMethod. This programme is specifically designed for Q-methodology (Schmolck, 2018) and supports identification of patterns among the different Q sorts. The programme provides a quantitative analysis to capture the common essence of the individual Q sorts into so-called 'typical' Q sorts. These follow a unique statement collection structure reflecting particular perspectives (Brown, 1993). First, all individual Q sorts are correlated with each other resulting in a correlation matrix, from which initial factors are extracted. The extracted factors are typical arrangements of statements or 'social perspectives' produced by the analysis of the expressions of the participants (Webler et al., 2009). The extraction is similar to standard factor analysis ('R' methodology) in regular quantitative data analysis but, where in R the initial factors are often extracted by means of Principle Component Analysis (PCA), in Q methodology it is more common to use the so-called Centroid procedure (McKeown & Thomas, 2013). Since this study was, as mentioned earlier, aimed at illuminating six different perspectives and not focused on how many additional perspectives are to be extracted altogether, Browns' Centroid procedure was preferred over Horst's Centroid procedure (Schmolck, 2018). From this, six factors were selected for manual rotation. After multiple rounds of rotation, it became clear that it was not possible to have at least three participants to positively 'load' onto each perspective or that each participant loaded positively on at least one perspective while loading negatively at another. In fact, two factors remained so similar that it was decided to repeat the analysis procedure with five factors instead. The results of this procedure are shown in Tables 2 and 3. This 'solution', as presented in Table 2, accounted for 59% of the variance (see Table 3) and allowed for establishing five distinct perspectives on sustainable urban tourism development. Through combining the statements ranked the highest and the lowest (array positions of +6, +5, −5, and −6 respectively) in the typical Q sort representing a particular perspective and linking these to the reflections of participants for assigning statements to these positions, these perspectives can be summarized as follows: Perspective A: tourism as a cause of crisis The basic premise for this perspective is that tourism does not necessarily make a city a better place to live and that there is a fixed limit to the number of tourists that it can 'absorb'. Whereas economic growth resulting from urban tourism does not necessarily have to come at social and environmental costs, a  Airbnb, StayOkay and the such are a positive contribution to the city, because they promote interaction between residents and tourists −4 0 4 −2 −1 2 It is important that local residents and local businesses feel their opinions and suggestions are taken seriously by the local government. 2 4 4 2 5 3 To optimize involvement of local residents and business, a citizen feedback app should be promoted and used 0 3 1 −4 3 4 Local residents are crucial to making visitors feel welcome, and therefore they need to be involved in decision making about urban tourism 3 4 3 −5 4 5 Residents are currently not enough involved in decision making with respect to sustainable urban tourism 5 2 5 −6 1 6 Bottom-up decision making for sustainable urban tourism is impossible, because the issues are too complex −5 −5 0 6 −2 7 Decisions with respect to urban tourism should always involve an element of bottom-up decision making 3 5 3 −1 2 8 Residents are not knowledgeable enough to understand the benefits of urban tourism for their well-being −1 −1 0 −1 3 9 Because many different departments, management layers, and interest groups need to be involved, decision making is too slow. 2 1 0 2 3 10 Policy interventions are not needed to promote sustainable urban tourism, because our capitalist market system will solve this issue automatically −6 −6 −5 1 −5 11 Over the years, urban tourism has been governed poorly, which has reduced faith in local government 2 −3 3 0 0 12 Over the years, culture has become a less important reason for tourists to visit a city 0 −5 −4 −6 −2 13 The friendliness of the local residents is a key part of the attractiveness of cities 1 5 2 3 5 14 Long term success of urban tourism requires preserving the authenticity of the city 5 6 2 2 5 15 Sustainable urban tourism requires changes to the city's infrastructure, which makes it impossible to maintain some of the characteristics that attract tourists −2 −1 −1 4 0 16 In spite of their environmental disadvantages cruises can support sustainable urban tourism by spreading/ controlling visitor flows −2 −1 −3 4 1 17 The solution to dealing with overcrowding in tourism is to focus on spreading tourists around the city and its surroundings 3 3 −3 2 1 18 Urban tourism is a typical example of a vicious circle: more tourists lead to changes in the retail landscape and other facilities which attracts even more tourists and local residents which inevitably leads to overcrowding and conflicts and that same area becoming less attractive and liveable In an effort to save our planet, stakeholders of urban tourism should focus more on technological solutions 1 3 −1 1 −2 20 Given that tourism experiences can often not be automated makes sustainable developments of this sector more problematic −3 1 −2 1 −2 21 Technology cannot decrease the gap between rich and poor −3 1 −6 4 1 22 Urban tourism reduces the gap between rich and poor −4 −4 −2 3 −4 23 Common strategies towards more sustainable urban tourism are difficult to agree upon, because stakeholders are too focused on their own interests 1 3 5 0 0 24 Long-term strategies for sustainable urban tourism are hard to deliver, because the election cycle forces politicians to focus on short term issues 2 2 1 −5 2 25 Promotion of sustainable urban tourism needs to be based on a clear long-term vision for the city as a whole in a few decades and therefore needs to be based on forecasting and negotiations that involve all relevant stakeholders 6 5 4 3 6 26 We will never achieve real transformation towards more sustainable urban tourism without having a crisis first −1 −6 1 −3 −1 27 From an ecological perspective, urban tourism is better than non-urban tourism, because its impact is easier to control, and solutions are easier to implement 4 0 2 −3 −5 28 Sustainable urban tourism implies banning non-electric cars and coaches from the city centres 0 0 3 −1 −3 29 A city should focus on reducing air travel towards the city, rather than minimizing emissions within the city, because this makes a bigger contribution to overall reduction of CO 2 emissions −1 −4 6 −3 −5 30 Increased urban tourism will automatically decrease social cohesion in the city 1 −1 −2 0 −3 31 Tourism changes the retail landscape in a city, which makes it more like a theme park than a place to live in 4 −2 0 4 −1 32 Seasonality is a blessing for liveability of the city −3 0 0 0 −1 33 To protect liveability in specific parts of the city, other parts should be designated for tourism-even if this decreases liveability in those designated parts −2 −3 −3 −3 −1 34 Reducing CO 2 emissions is more important than improving living conditions for local residents −1 1 −4 −5 −4 35 Policy makers should focus more on environmental sustainability than on living conditions for local residents 0 −3 −5 −2 −3 36 Ultimately, liveability of the city is more important than its appeal to tourists. 4 0 6 −2 2 (Continued ) 38 Cities should focus on attracting smaller numbers of high spending tourists rather than large numbers of tourists that spend less −1 −1 0 1 0 39 To save the planet, the number of tourists needs to decrease −1 −2 0 −4 −6 40 The success of urban tourism should never be expressed in number of tourists 1 4 5 −4 4 41 There is a fixed limit to the number of tourists a city can absorb 6 1 −4 1 0 42 Urban tourism should be treated as a business case. In other words, decisions need to be based on return on investment −1 −3 −1 1 3 43 Business travellers are more inclined to prefer sustainable alternatives, also as a result of company policies. Therefore, cities that want to stimulate sustainable urban tourism need to focus more on this market segment to get the ball rolling 0 −1 1 0 −1 44 To become sustainable, a city needs to attract more highly educated and talented people as residents −4 1 −5 3 −4 45 Tourism makes a city a better place to live in −6 2 2 2 1 46 Tourism is more beneficial for tourists than for local residents −2 −2 2 −1 −3 47 Tourism helps to maintain the cultural heritage of the city 2 3 −1 0 6 48 Sustainable urban tourism depends on communicating to tourists what kind of behaviour is expected from them and what kind of behaviour is unacceptable 0 0 −1 5 1 49 Because tourists are less involved with the city, the only way to influence their behaviour is through enforcement −2 −2 −6 3 0 50 If you don't like tourists, you should not live in a city −5 0 −1 0 1 51 Promoting your city to tourists based on sustainability is a mission impossible −4 −5 −3 −1 −4 52 Too much money has been invested in promoting the city, instead of promoting the well-being of its residents 3 −3 0 −1 −3 53 Residents should applaud cities for investments to attract more tourists because this makes cities more attractive places to live in −3 2 −2 0 4 54 The short-term focus of urban tourism harms its potential for long-term economic growth 5 −1 1 −3 0 55 Promoting environmental sustainability requires economic incentives 4 2 3 −2 4 56 Economic growth of urban tourism can only come at social and environmental costs −5 −4 1 1 −6 57 Economic taxation and fines do not lead to sustainable urban tourism 0 0 −2 −1 0 58 The economic contribution of the residents is just as important as the contribution of the tourists 1 4 −4 5 2 59 Sustainable development of urban tourism leads to cost savings for local stakeholders −3 1 −1 −2 2 60 The long-term economic success of cities depends as much on pleasing local residents as it does on attracting as many visitors as possible 0 6 4 −4 3 short-term focus harms its potential to contribute to long-term economic growth. Therefore, planning aimed at realizing sustainable urban tourism planning needs to be based on forecasting and a clear long-term vision for the city as a whole. Negotiations about actual planning decisions need to be based on these forecasts and this vision and require involvement of all relevant stakeholders. A key reference point for this perspective is that residents are currently not sufficiently involved in decision-making. Long-term success of sustainable urban tourism requires preserving the authenticity of the city and doing so requires policy interventionsbased on a long-term vision, forecasts and stakeholder/resident involvementbecause leaving this issue to 'the market' will not solve the problem.

Perspective B: tourism as a manageable sector
This perspective supports the necessity to involve residents in decision-making regarding urban tourism planning and is equally unimpressed by the potential of a 'laissez-faire' approach to realizing sustainable urban tourism. However, whereas perspective A could probably be described as interpreting the current situation as a crisis, perspective B relates to more optimism about the current situation. In fact, a key reference point for this perspective is that a transformation of urban tourism in a sustainable direction could very well be realized without having (the increased sense of urgency associated with) going through a crisis first. However, this requires putting more emphasis on the role and needs of residents, given that long-term economic success of cities depends as much on 'pleasing' their residents as it does on attracting tourists. In fact, those same residentsand the way they respond to and welcome touristsare a key part of the attractiveness of the city for tourists. Therefore, any urban tourism planning decision should involve residents and an element of bottomup decision-making.
Perspective C: technology as a means to save the day Similar to perspectives A and B, perspective C acknowledges the value of involving residents in decision-making regarding urban tourism planning and that tourism development cannot be left to our capitalist free market system. What is more, the reference point for this perspective is that liveability of the city should trump its appeal to tourists in decision-making and puts even more emphasis on this than perspective A while outscoring perspective B in relation to stressing that the success of urban tourism should never be expressed in terms of the number of tourists. Very much in contrast to perspective A though, this perspective is based on the assumption that there is no fixed limit to the number of tourists that a city can 'absorb'. Ultimately, this perspective mostly defines success of sustainable urban tourism in terms of reducing the gap between rich and poor, as well as reducing CO 2 emissions through reducing air travel. In fact, the premise of this perspective is that cities should focus on the latter instead of trying to minimize emissions within the city itself. These issues cannot be resolved through attracting more talented and highly educated residents or enforcement but require technological breakthroughs. In other words, this perspective stresses the need to account for the interests of all stakeholdersincluding the environmentbut ultimate solutions will have to be supported by technology rather than trying to agree on common strategies towards more sustainable urban tourism with all stakeholders because most stakeholders are simply too focused on their own interests to make this work.

Perspective D: tourism in need of strong leadership
Perspective D is based on a rather pessimistic view on bottom-up decision making because it assesses the issues at hand as simply too complex. In fact, this perspective is based on the reference point that local residents should thus not be involved in decision-making with respect to realizing sustainable urban tourism. This perspective views urban tourism as a typical example of a vicious circle: more tourists results in changes in the retail landscape and other facilities, which attracts even more tourists and residents to those areas, which inevitably leads to overcrowding and conflicts and, ultimately, those same areas becoming less attractive to visit and live. This perspective is the only one to firmly disagree with the statement that election cycles force politicians to focus on the short term and that this makes escaping this vicious circle even more difficult. It is also the only perspective to not firmly disagree with the statement that our capitalist free market system will solve some of the current issues automatically. Out of all perspectives, perspective D is the one that puts most emphasis on the negative impacts of developments such as Airbnb and the importance of communicating to tourists about the types of behaviour that are expected of them in trying to mitigate these impacts. In other words, this perspective is based on the firm belief that policymakers are the ones that should and could define and implement effective policies to stimulate sustainable urban tourism, while possibly leaving some issues to be resolved by 'the market'.

Perspective E: tourism as a force for good
This final perspective is based on the premise that tourism can actually help to maintain the cultural heritage of a city. Out of all perspectives, this perspective is also the one to most firmly disagree with the need to reduce the number of tourists to save our planet. In fact, this perspective is explicitly based on the reference point that economic growth associated with urban tourism need not come at the expense of negative social and environmental impacts. This perspective is very much based on rejecting the idea that air travel needs to be reduced. Its premise is that sustainable urban tourism can actually be realized through a clear long-term vision and making sure that local residents and businesses feel that policymakers take their opinions and suggestions seriously. In other words, out of all perspectives, this one is by far the most positive and optimistic about both the current and future state of urban tourism.
The 'loadings' presented in Table 3 represent the degree to which an individual Q-sort of an individual participant correlates to a particular perspective (typical Q-sorts). These can be positive or negative and basically indicate the degree to which a particular participant has affinity with a particular perspective. For loadings to reach significance of p < .05, they must exceed ±.253 (1.96*(1/√N ) with N = number of statements (=60)) (McKeown & Thomas, 2013). Table 3 shows that the participating 'residents' both significantly positively loaded onto perspective B. However, the 'resident' from Goteborg also positively loaded onto perspective C whereas the 'resident' from Darmstadt did so on perspective E. Both showed to be indifferent towards perspective A or D. From the participants representing 'city marketing', two (one from Darmstadt and one from Stavanger) seemed to have similar patterns (both having positive loadings onto perspective B and E, and the participant from Stavanger also positively loaded onto perspective D). The other two (both from Goteborg), however, showed different patterns: one participant positively loaded onto A and C, whereas the other positively loaded onto C and E and actually negatively loaded onto B. The eight participating 'entrepreneurs' all showed to have different loading patterns. Almost all (seven) positively loaded onto B, five did so onto D, four onto A, three onto C, and three onto E. Remarkably, the one participant that did not positively load onto B, is the sole participant to negatively load onto C. This participant (from Amsterdam) does share positive loadings onto perspectives D and A, as most of the other 'entrepreneurs'. Both 'NGO representatives' showed to have very similar loading patterns; both significantly positively loaded onto A and E.
Two of the six 'policymakers' (both from Valencia), had similar loading patterns (both only significantly loaded onto perspective A) but, in fact, that seemed to be the only similarity among the 'policymakers'. To illustrate this, consider for example the fact that the 'policymakers' from Stavanger and Amsterdam both positively loaded onto B, but with the participant from Stavanger additionally positively loading onto C and the participant from Amsterdam loading onto E. A more conflicting point of view was found for one 'policymaker' (from Valencia); this participant positively loaded onto D, but also significantly negatively loaded onto A and B. This pattern remarkably showed great similarities with the pattern of the 'academic' (Valencia), who also significantly negatively loaded onto A and B, and significantly positively loaded onto D. But, in addition, the 'academic' also negatively loaded onto E, whereas the 'policymaker' did not. Table 3 thus highlights that, apart from similar patterns found among the three participants from Darmstadt and among the two 'NGO representatives', no other similarities between patterns over the cities or over the 'roles' could be established. Therefore, the five different perspectives identified in this study actually seem to be rather 'randomly' divided over the participants, in contrast to the initial assumption on which the initial expectation of being able to distinguish six distinct perspective was based. Indeed, contrasting the strongest and lowest loadings of the participants (Table 4) illustrates even more clearly the strong differences between people with similar roles in tourism.

Discussion
The current paper has clearly brought to the foreground the diversity of worldviews among relevant stakeholders related to (sustainable) urban tourism development. The identification of five dominant worldviews helps to further understanding on the way overtourism and the way it should be governed is perceived by stakeholders. The binary emphasis in the popular debate between proponents and adversaries of tourism may drown out other more nuanced worldviews, which are, by their very nature, diverse and complex. Current collaborative planning efforts based on the principles of seeking consensus and creating shared perspectives can therefore cause underlying concerns of people to be ignored, which can create resentment towards potentially useful solutions instead of support for those solutions, while also preventing relevant ideas and solutions from being suggested and discussed altogether. In contrast, distilling a wide variety of perspectives into coherent worldviews, as shown in this paper, can support truly participative planning by acting as alternative ways of framing the debate and explicitly giving voice to perspectives that are currently drowned out. This can facilitate a different dialogue on tourism development that may very well prove more productive than the current ones and provide opportunities for innovative insights and solutions being incorporated in strategies for dealing with this wicked problem.
More generally, if we accept that sustainable urban tourism development strategies are driven and informed at least in part by individual worldviews, it may actually be impossible to make 'objective' policy choices. It might be more useful instead, to explore possibilities to use the diversity of worldviews to enrich the discussion on tourism development. This will allow stakeholders to reflect on the values of themselves and others and create a better understanding of what sustainable tourism development entails for different people at different places and moments in time, given that this may very well change as the situation changes. This line of reasoning fits with the discussions on creating more resilient destinations (Saarinen & Gill, 2018), and even provides a practical way to stimulate resilient thinking among stakeholders. However, to allow this kind of resilience thinking to take place will require new governance arrangements; arrangements that emphasize empathy for disagreeing opinions and a resulting reflectivity on one's own perspective, for instance those suggested in transition management literature (e.g. Frantzeskaki et al., 2017), instead of regarding consensus-seeking as the 'holy grail of tourism governance'.
Our findings refute the implicit assumption in previous debates on (urban) tourism development and (urban) tourism governance, as well as the underlying logic for many current practices towards realizing collaboration and stakeholder involvement, that there is some sense of uniformity among people's perspective on and their (main) role or function within the (urban) tourism ecosystem (e.g. Byrd, 2007;Dodds, 2007;Jamal & Stronza, 2009;Maxim, 2019;Ritchie & Crouch, 2003;Wang & Fesenmaier, 2007;Zehrer & Hallmann, 2015). Not only do the findings confirm the diversity in perspectives among individual stakeholders (Beritelli, 2011a;Pforr, 2006;Scott et al., 2008), they also highlight how there is little consistency/homogeneity in perspectives of people with similar roles within one and the same destination. For instance, Table 3 shows that within the 'category' of 'entrepreneurs' both perspective A and E are present; one representing the most pessimistic perspective on the current and future situation regarding urban tourism and the other the most optimistic. What is quite striking is that the same applies to 'policymakers'. In fact, Table 4 also shows that within one and the same destination, you could have policymakers that value bottom-up decision-making and its contribution to formulating appropriate sustainable urban tourism development strategies and policymakers that wholeheartedly disagree with that train of thought and feel that policymaking should be left to (professional) policymakers.
These are just a few examples of how the results show that the relation between relevant stakeholders' personal worldviews and their (main) roles or functions, even within the same destination, is A* E** Asterisks indicate significant loadings on that specific perspective at *p < 0.05 or **p < 0.01.
far from uniform. This may go some way to explain why it is so difficult to change the growth paradigm and policy discourses aimed at tourism growth, even in cities where there is strong public pressure to limit the increase of tourism. The diversity of individual worldviews, also among people with similar roles in tourism, limits opportunities for generating new consensus-based collective worldviews. It may very well prove impossible to collaboratively and jointly formulate 'new' policies that please 'everyone at the table'. Consequently, the 'old'institutionalizedcollective worldview is likely to remain the reference point for 'new' sustainable urban tourism development strategies.
To illustrate this point further, consider the findings of this research regarding entrepreneur 8, policymaker 6 and academic 1 all agreeing with worldview A. The basic premise for this worldview is that tourism can have highly negative impacts and that there is a fixed limit to the number of tourists that a city can 'absorb'. They all disagree with worldview E, which advocates tourism growth. It is interesting to find that entrepreneurs too hold this worldview, as this would suggest possibilities for consensus-seeking collaborative approaches. However, resident 2 and policymaker 2 display the exact opposite perspectives in that they are in favour of tourism growth and oppose regulation of tourism numbers. It will be very difficult to bring these two groups together and one must wonder if it would be at all possible, or even desirable, to bring all of these stakeholders together to come to a shared vision if this means limiting the possibilities for conflicting perspectives (Beritelli, 2011b). Instead, it may be more beneficial to reframe the ultimate debate, using one or more of the other worldviews, to come to innovative insights and governance perspectives. To facilitate such a debate, it may prove useful to turn towards governance frameworks that are designed to facilitate debate and collaborative reflections on tourism development in a systemic way (e.g. Koens, Melissen, Mayer and Aall, 2019). The five worldviews that are recognized in this paper could be used to provide alternative interpretations and help build bridges between with different perspectives, no matter what organization or role involved stakeholders represent. The focus here may very well have to be on joint understandings, rather than consensus.

Limitations and ideas for future research
This research study has sought to provide insights with regards to the ways in which worldviews can influence participatory planning governance. It is important to recognize though, that primary data gathering has taken place in a European context. Even though there is no clear distinction between the different cities, this is not to say these worldviews are universal. For example, it is highly likely that different worldviews exist in, for example, the Global South, where the opportunities and challenges for tourism development and governance are very different. There are plenty of cities where there still is room for tourism to grow and contribute to local and regional development. In places where this is the case, it seems plausible or even likely that this will result in different collective and individual worldviews. As such, it would be interesting to replicate the study elsewhere, where the overarching narrative on tourism is not as polarized as it is in large parts of Europe at the moment.
Also, whilst the recognition of five different worldviews can shed some light on the difficulties in facilitating collaborative planning, it was beyond the scope of the research to investigate to what extent certain worldviews align with, for example, respondents' political alignment, their ideological views on the general regulation of economic systems or their demographic characteristics. Doing so could, however, provide deeper insights and understanding into the perspectives of individual stakeholders.
Looking towards future research, it would be interesting to look at ways to accommodate a participatory planning process that takes into account these different worldviews. Given that it is impossible to expect a unanimous point of view among stakeholders and even individual members within a specific stakeholder group, it can be useful to investigate alternative ways of participatory planning. Already, destinations are experimenting with innovative and experimental governance approaches, for example through the development of living labs that allow stakeholders to design, test and learn about governance innovations in real time (von Wirth, Fuenfschilling, Frantzeskaki, & Coenen, 2019). It would be useful to use such labs to also experiment with alternative participatory planning approaches as part of these governance processes.
Simultaneously, as pointed out earlier, the body of work dedicated to exploring and understanding the complexities involved with recognizing, accounting for and actually benefitting from varying stakeholder perspectives on destination development and management in tourism governance, especially linked to the specific context of sustainable urban tourism development and dealing with overtourism, is still rather limited. Therefore, academic perspectives on urban tourism governance and application on the ground would surely benefit from further research into this topic, also to verify and expand on the findings, and their implications, presented in this paper.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work is part of the research programme JPI Urban Europe -ERA-NET Cofund Smart Cities and Communities with project number 5619886, which is (partly) financed by the Netherlands Organisation for Scientific Research (NWO).